public inbox for linux-media@vger.kernel.org
 help / color / mirror / Atom feed
From: "Richard Röjfors" <richard.rojfors@pelagicore.com>
To: Pawel Osciak <p.osciak@samsung.com>
Cc: 'Linux Media Mailing List' <linux-media@vger.kernel.org>,
	'Mauro Carvalho Chehab' <mchehab@redhat.com>,
	'Douglas Schilling Landgraf' <dougsland@gmail.com>,
	'Samuel Ortiz' <sameo@linux.intel.com>
Subject: Re: [PATCH 1/3 v2] media: Add a cached version of the contiguous video buffers
Date: Wed, 04 Aug 2010 11:40:22 +0200	[thread overview]
Message-ID: <4C593586.6030804@pelagicore.com> (raw)
In-Reply-To: <000d01cb33aa$606faee0$214f0ca0$%osciak@samsung.com>

On 08/04/2010 09:55 AM, Pawel Osciak wrote:
> Hi Richard,
>
>> Richard Röjfors wrote:
>> This patch adds another init functions in the videobuf-dma-contig
>> which is named _cached in the end. It creates a buffer factory
>> which allocates buffers using kmalloc and the buffers are cached.
>>
>
> Before I review this in more detail, could you elaborate more on
> this? How large are your buffers, can kmalloc really allocate them
> for you? I am not convinced how this is supposed to work reliably,
> especially in a long-running systems.

The buffers are normally 829440 bytes and yes kmalloc can allocate them.
Normally userspace apps seem to request two buffers of this size.

How do you propose to allocate the buffers? They need to be contiguous
and using uncached memory gave really bad performance.

--Richard

  reply	other threads:[~2010-08-04  9:40 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-03 15:18 [PATCH 1/3 v2] media: Add a cached version of the contiguous video buffers Richard Röjfors
2010-08-04  7:55 ` Pawel Osciak
2010-08-04  9:40   ` Richard Röjfors [this message]
2010-08-04  9:50     ` Pawel Osciak
2010-08-04 10:03       ` Richard Röjfors
2010-08-04 10:34         ` Pawel Osciak
2010-08-04 11:37           ` Richard Röjfors
2010-08-05 20:25           ` Richard Röjfors
2010-08-09  8:50             ` Pawel Osciak

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4C593586.6030804@pelagicore.com \
    --to=richard.rojfors@pelagicore.com \
    --cc=dougsland@gmail.com \
    --cc=linux-media@vger.kernel.org \
    --cc=mchehab@redhat.com \
    --cc=p.osciak@samsung.com \
    --cc=sameo@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox